DGX Spark (Grace Blackwell GB10) Performance Drop: GPU Trapped in 15W / 650MHz Loop with 50°C Artificial T.Limit Temp

I am experiencing a severe, sudden performance drop on my NVIDIA DGX Spark system where the integrated Blackwell GPU gets trapped in an emergency low-power state. Even under 99% compute utilization workloads, the card refuses to scale past 15W and locks its graphics frequencies to 650 MHz.

The issue is accompanied by a massive accumulation of microseconds in the SW Power Capping log counters and an artificially lowered GPU T.Limit Temp threshold pinned exactly at 50°C (despite the chip running cool at 44°C).

I have attached a Grafana telemetry snapshot showing the sudden power cliff profile, followed by the relevant nvidia-smi diagnostic logs below.

System Environment:

  • Hardware: NVIDIA DGX Spark (Integrated GB10 Grace Blackwell SoC)

  • Driver Version: 580.142

  • CUDA Version: 13.0


Diagnostic Logs (nvidia-smi Output):

1. Clock Throttling Evidence (Locked at 650 MHz instead of 2418 MHz default):

Timestamp                                              : Sat May 16 02:28:57 2026
Driver Version                                         : 580.142
CUDA Version                                           : 13.0

Attached GPUs                                          : 1
GPU 0000000F:01:00.0
    Clocks
        Graphics                                       : 650 MHz
        SM                                             : 650 MHz
        Memory                                         : N/A
        Video                                          : 598 MHz
    Applications Clocks
        Graphics                                       : 2418 MHz
        Memory                                         : N/A
    Default Applications Clocks
        Graphics                                       : 2418 MHz
        Memory                                         : N/A
    Deferred Clocks
        Memory                                         : N/A
    Max Clocks
        Graphics                                       : 3003 MHz
        SM                                             : 3003 MHz
        Memory                                         : N/A
        Video                                          : 3003 MHz
    Max Customer Boost Clocks
        Graphics                                       : N/A
    SM Clock Samples
        Duration                                       : N/A
        Number of Samples                              : N/A
        Max                                            : N/A
        Min                                            : N/A
        Avg                                            : N/A
    Memory Clock Samples
        Duration                                       : N/A
        Number of Samples                              : N/A
        Max                                            : N/A
        Min                                            : N/A
        Avg                                            : N/A
    Clock Policy
        Auto Boost                                     : N/A
        Auto Boost Default                             : N/A

2. Temperature Lock (Artificial 50°C T.Limit):

Timestamp                                              : Sat May 16 02:29:48 2026
Driver Version                                         : 580.142
CUDA Version                                           : 13.0

Attached GPUs                                          : 1
GPU 0000000F:01:00.0
    Temperature
        GPU Current Temp                               : 44 C
        GPU T.Limit Temp                               : 50 C
        GPU Shutdown T.Limit Temp                      : N/A
        GPU Slowdown T.Limit Temp                      : N/A
        GPU Max Operating T.Limit Temp                 : 0 C
        GPU Target Temperature                         : N/A
        Memory Current Temp                            : N/A
        Memory Max Operating T.Limit Temp              : N/A

3. Performance Counters (1,369+ Seconds of Active SW Power Capping Loop):

Timestamp                                              : Sat May 16 02:31:03 2026
Driver Version                                         : 580.142
CUDA Version                                           : 13.0

Attached GPUs                                          : 1
GPU 0000000F:01:00.0
    Performance State                                  : P0
    Clocks Event Reasons
        Idle                                           : Not Active
        Applications Clocks Setting                    : Not Active
        SW Power Cap                                   : Not Active
        HW Slowdown                                    : Not Active
            HW Thermal Slowdown                        : Not Active
            HW Power Brake Slowdown                    : Not Active
        Sync Boost                                     : Not Active
        SW Thermal Slowdown                            : Not Active
        Display Clock Setting                          : Not Active
    Clocks Event Reasons Counters
        SW Power Capping                               : 1369386233 us
        Sync Boost                                     : 0 us
        SW Thermal Slowdown                            : 0 us
        HW Thermal Slowdown                            : 0 us
        HW Power Braking                               : 0 us
    Sparse Operation Mode                              : N/A

4. Power Readings (Hard-capped at ~15W):

Timestamp                                              : Sat May 16 02:28:39 2026
Driver Version                                         : 580.142
CUDA Version                                           : 13.0

Attached GPUs                                          : 1
GPU 0000000F:01:00.0
    GPU Power Readings
        Average Power Draw                             : 14.11 W
        Instantaneous Power Draw                       : 15.36 W
        Current Power Limit                            : N/A
        Requested Power Limit                          : N/A
        Default Power Limit                            : N/A
        Min Power Limit                                : N/A
        Max Power Limit                                : N/A
    Power Samples
        Duration                                       : Not Found
        Number of Samples                              : Not Found
        Max                                            : Not Found
        Min                                            : Not Found
        Avg                                            : Not Found
    GPU Memory Power Readings 
        Average Power Draw                             : N/A
        Instantaneous Power Draw                       : N/A
    Module Power Readings
        Average Power Draw                             : N/A
        Instantaneous Power Draw                       : N/A
        Current Power Limit                            : N/A
        Requested Power Limit                          : N/A
        Default Power Limit                            : N/A
        Min Power Limit                                : N/A
        Max Power Limit                                : N/A


What I Have Tried:

  1. Verified that the device is connected strictly via the official heavy-duty NVIDIA power brick directly to a reliable wall source to rule out power rail voltage sags. Validated socket.

  2. nvidia-powerd.service is not present/loaded on this specific server distribution setup.

Is this an emerging bug in Driver 580.142/CUDA 13 for the Blackwell PMIC architecture, or is there a low-level hardware firmware lock/register reset mechanism I need to trigger via the DGX container stack to release this safety state?

Any guidance from the engineering team would be highly appreciated.


Try fully unplugging it from the wall for 10 minutes I think that may be able to solve it.

I think 1 or 2 minuets will work.

You can run this program to check

TL;DR - it’s not clear whether you were putting the system under load when you took those nvidia-smi metrics. But I would like to clarify some points.

GPU T.Limit Temp in itself at 50 degrees is OK, it means that you have 50 degrees of overhead available before target temperature is reached. If you put a workload on your GPU then you should see this value reduce. Are you saying it doesn’t reduce under load?

As for the clocks, once again, here is some output from mine:

    Clocks
        Graphics                                       : 208 MHz
        SM                                             : 208 MHz
        Memory                                         : N/A
        Video                                          : 598 MHz
    Applications Clocks
        Graphics                                       : 2418 MHz
        Memory                                         : N/A

This is normal, the system isn’t under load currently. Once I load a model, it starts to increase:

    Clocks
        Graphics                                       : 910 MHz
        SM                                             : 910 MHz
        Memory                                         : N/A
        Video                                          : 1053 MHz
    Applications Clocks
        Graphics                                       : 2418 MHz
        Memory                                         : N/A

And a little later:

    Clocks
        Graphics                                       : 2457 MHz
        SM                                             : 2457 MHz
        Memory                                         : N/A
        Video                                          : 2138 MHz
    Applications Clocks
        Graphics                                       : 2418 MHz
        Memory                                         : N/A

Now, SW Power Capping - your value of 1369386233 us. This is in microseconds, so we need to divide by 1 million to see the value in seconds = 1369 seconds. This means that the system was under load since the last reboot, and has spent a total of 1369 seconds in a power cap mode. What is more important is “SW Power Cap” is not active, which means that the system is not currently power capped. I also have non-zero values for this, and I have found this increments at times of particularly high load (90W GPU draw plus).

Isn’t the real issue here the Applications Clocks Graphics: 2418 MHz? I noticed that my clock speed fell back to this as well after a full-on crash and I lost 20% performance compared to my earlier benchmark results with ASUS Ascent GX10.

i have exactly the same problem, i have two GB10 from gygabite one has the issue the other does not.

Could you try to swap their power supplies (and/or cool them with additional fan)? I have strong suspicion this is kind of power supply problem. Overheating or damaged cable/connector causes voltage drop and spark’s power controller switches to 45 or 65W USB-C PD power profile.

This is similar to other issues that are under investigation. Unplugging the unit AND the power supply from the wall should bring the unit back to normal operation